Abstract A clinic-based translation of the Diabetes Prevention Program (DPP) tailored to Black and Latino men has the potential to address prior limitations of DPP implementation and reduce diabetes disparities among men. The proposed study is designed to 1) Assess the effect of Power-Up vs. standard care National DPP (NDPP) on percent weight loss among racial/ethnic minority men at risk for diabetes; 2) Compare engagement of racial/ethnic minority men at risk for diabetes in Power-Up vs. standard care NDPP; and 3) Evaluate the Reach, Effectiveness, Adoption, Implementation, and Costs of Power-Up using the RE-AIM framework. We hypothesize that men randomized to Power-Up will achieve significantly greater weight loss (% weight loss from baseline) at 16-weeks and 1-year than men randomized to the standard care, mixed-gender NDPP group. Men randomized to Power-Up will also have significantly greater engagement rates (i.e., attend at least 4 sessions) and retention rates (i.e., attend 9 or more sessions) than men randomized to the standard care NDPP. Using the electronic health record (EHR) systems of a large academic medical center and a network of small to medium independent primary care practices throughout New York City, we will identify Black and Latino men that meet the BMI and A1c eligibility requirements for NDPP. We expect to enroll 450 participants through our health system partners. Men will be randomized 1:1 to either the Power-Up intervention arm or referred to the standard NDPP at their clinic site. Consistent with current CDC standards and current NDPP practices of our health system partners, telephone make-up sessions will be offered by coaches in both conditions to men who miss in-person sessions. The primary outcome will be percent weight loss over 12 months, measured at baseline, approximately 16 weeks (end of core sessions), and 12 months (end of maintenance sessions). Study-collected weight will be the primary outcome measured at the first Power-Up or NDPP session by coaches (or from study staff or clinic sites when sessions are not attended). With a sample size of 450 men, a standard deviation of 4.7 percentage points in percent weight loss, and an intraclass (intra program group) correlation of 0.05 we would have 98.2% power to detect a 2.5 percentage point difference between the intervention arms at the 0.05 significance level. This sample size would also give us 80% power to detect a 13.8 percentage point difference in engagement, and a 16.4 percentage point difference in retention. Tertiary outcomes related to the RE-AIM evaluation include quantitative and qualitative data collected via validated self- report questionnaires administered via telephone survey, EHR-data extraction, and qualitative interviewing of individuals and focus groups. Our evaluation will answer questions most important to healthcare systems and NDPP providers and our partnerships with diverse healthcare systems will allow us to disseminate the intervention broadly, if proven effective.